BIOCHIMICAET BIOPHYSICAACTA
252
STUDIES
ON PHOSPHOLIPASE
A AND ITS ZYMOGEN
FROM
PORCINE
PANCREAS
III. ACTION OF THE ENZYME
ON SHORT-CHAIN
G. H. DE HAAS, P. P. M. BONSEN, W. A. PIETERSON
Laboratory
of Biochemistry,
The State University,
LECITHINS
AiTD I,.
Vondellaan
2 6,
L. M. VAN DEENEN
Utvecht
(The
Netherlands)
(Received March z6th, 1971)
SUMMARY
I. Short-chain lecithins (with Cg, C,, and C, fatty acid esters) have been used
to study kinetically the enzymatic hydrolysis by pancreatic phospholipase A (EC
3.1.1.4) in aqueous systems, without the addition of emulsifiers.
2. Although phospholipase A is able to attack these substrates in molecularly
dispersed form, micellar solutions are hydrolyzed at a much higher rate.
3. Of the three substrates examined, dioctanoyllecithin appeared to be the
best substrate. Differences in maximal velocities might be interpreted in terms of
interfacial area per molecule.
4. Caz+ is specifically required for activity of pancreatic phospholipase A. The
kinetic results are consistent with a random mechanism in which the metal ion combines with the enzyme independently of the substrate. The substrate was found to
combine with the enzyme independently of the metal ion concentration.
5. Kinetic parameters were determined with diheptanoyllecithin as a substrate
over a pH range from 5 to g. Maximal binding of enzyme with substrate was observed
at pH < 6. The affinity of the enzyme for Ca 2+ decreased at pH values below 6.5.
6. VVith diheptanoyIlecithin as substrate, maximal velocities at infinite substrate and Ca2+ concentrations showed an optimum at pH 5.75.
7. NaCI at high concentrations (up to 3.9 &I) gave a So-fold stimulation of the
?I,,, (diheptanoyllecithin as substrate). The KS value decreased slightly with increasing salt concentrations, while the Kca2+ increased very strongly. The activating
effect of salt is presumed to be caused by a change of the properties of the lipidwater interface.
INTRODUCTION
Phospholipase A (EC 3.1.x.4) catalyzes the specific hydrolysis of fatty acid
ester bonds at the 2-position of x,z-diacyl-sn-phosphoglyceridesl.
A pH optimum in
Abbreviation:
Biochim.
CMC,critical micelle concentration.
Biophys.
Acta,
239 (1971)
252-266
PHOSPHOLIPASE
A
2.53
alkaline solution is reported generally, as is a requirement of Ca2+ for activity. This
wide-spread enzyme has been purified from various snake venomsF1’, bee venom12p13,
and from pancreatic tissues14-17. Characteristic properties of the enzyme are its high
heat stability, its low molecular weight, and a compact structure caused by a high
number of disulfide bridges. In the pancreas of man’*, ratls, and pig15 the enzyme does
not occur in the active form, but is secreted as an enzymatically inactive zymogen.
Limited proteolysis by trypsin produces the active enzyme. Recently, the complete
amino acid sequence and the position of the disulfide bridges has been reported for
the porcine pancreatic phospholipase A and its zymogen20p21. SAMEJIMA et a1.22,
reported the partial sequence of a snake venom phospholipase A (Agkistrodon
halys
blomhojii).
In general, phospholipase A is not able to hydrolyze aqueous dispersions of
long-chain lecithins. The lecithins are insoluble in water, but they produce stable
dispersions by forming large lamellar structures, called liposomes. Such systems consist of closed, concentric lipid bilayers, separated from each other by water layers.
Dependent on their chemical structure (degree of unsaturation, chain length) the
lecithin molecules are more or less closely packed.
Addition of bile salts break the tightly packed lamellar structure down to much
smaller mixed micellesZ3, and a rapid hydrolysis of the lecithin molecules by phospholipase A occursle. Also the addition of diethylether24 or acidic phospholipid+, which
might introduce more space between the phospholipid molecules, can transform the
inert liposomes into more loosely packed structures which are susceptible to phospholipase A attack.
It seems of interest to investigate more closely how the action of the enzyme
might be controlled by specific properties of the lipid-water interface, such as area
per phospholipid molecule, charge effects, etc. Unfortunately, incubation systems
containing organic solvents or bile salts are too complicated to be useful for kinetic
analysis of the enzyme.
To overcome these problems, ROHOLT AND SCHLAMOWITZ~~
proposed already
in 1961 the use of the water-soluble dihexanoyllecithin as substrate. They were able
to show that crude phospholipase A from Crotalus durissimus terrijicus attacks this
substrate in molecularly dispersed aqueous solution without the addition of detergents.
The present study is an attempt to further characterize the purified pancreatic
enzyme with several synthetic short-chain 3-sn-phosphatidylcholines as substrates.
Dependent on fatty acid chain length and lipid concentration, these lecithins give
either molecularly dispersed, micellar or lamellar structures in water, and it seemed of
interest to investigate the kinetic parameters of the enzyme as a function of such
different lipid-water interfaces, and in addition as a function of metal ions, ionic
strength, and pH.
MATERIALSANDMETHODS
Enzyme source
Porcine pancreatic phospholipase A,* was obtained by the activation
pure zymogen with trypsin as described previously15.
* This enzyme will be denoted as phospholipase
A, unless otherwise
Biochim.
Biophys.
of the
stated.
Acta.
239
(1971)
252-266
G.
254
H. DE HAAS et at.
Substrates
3-sn-Phosphatidylcholines
with two identical acylchains varying in chain
length between pentanoic acid (C,) and nonanoic acid (C,) * were prepared as described by CUBERO ROBLES et al.z7~zs.
Assays
Enzymic hydrolysis of the synthetic lecithins was followed by continuous
titration of the liberated fatty acids at constant pH (between pH 5 and IO) at 40”
with a Radiometer pH-stat equipment. The assay-system was standardized to 4 ml
total volume and contained lecithin, CaCl, and NaCl in various amounts. Depending
on the pH, 0.5 mM Tris or acetate buffer was used. Lyophilized enzyme, dissolved in
distilled water, was added in aliquots of 10-100 ~1. Titrations with 0.04 M NaOH were
carried out under nitrogen. In the pH range used, non-enzymatic hydrolysis of substrate was negligible. Initial rates of hydrolysis were measured under conditions where
a good proportionality between enzyme concentration and activity existed. Analysis
of the reaction products showed only the formation of a lysolecithin and fatty acid.
At pH values below 5.0, the pH-stat technique cannot be used to measure enzyme activity, and the hydroxamate method as modified by AUGUSTYNANDELLIOTT@
was applied. Specific activity of the enzyme is expressed as the number of ,uMoles
of fatty acid released per min per mg of protein (dry weight).
Titration eficiency
Between pH 5.0 and 7.0 corrections have to be made for the titration efficiency
of C,-C, fatty acid. Fig. I shows the percentage of fatty acid which is titrated under
the average assay conditions. This pattern is independent of the Ca2+ concentration.
At pH 7.0 the fatty acids are virtually completely ionized.
Critical micelle concentration determinations
Determinations of the critical micelle concentrations (CMC) were carried out by
measuring the spectral shift induced upon incorporation of Rhodamine 6G into
micelies, as described by BECHER~@.
RESULTS AND DISCUSSION
Activity of@zospholi$ase
A with short-chain lecithins as substrates
The activity of phospholipase A was examined with short-chain lecithins. These
lecithins (diheptanoyllecithin and lower homologues) give optically clear solutions in
water and no saturation point has been found up to concentrations of about IOO
mg lecithin per ml. Micelle formation, however, occurs at low concentrations for dihexanoyllecithin and diheptanoyll~cithin (13.8 and z mM, respectively**).
As shown in Fig. 2, micellar solutions of, for instance diheptanoy~ecithin, are
readily hydrolyzed by the pancreatic phospholipase A in the absence of detergents
* According to the tentative rules for lipid nomenclature (IUPAC) the preferred name for these
compounds is: r,z-diacyl-sn-glycero-3-phosphorylcholines.
For the sake of simplicity these will
be abbreviated to dipentanoyllecithin,
e2c.
* + These values were confirmed by the light-scattering technique and by surface tension measurements.
Biochina. ~~ophys. Acta, 239
(1971)
252-266
PHOSPHOLIPASE
A
255
LO
6
%120
ET
1100
o-o-o-o-o-o-o
/d
g
S80
60
40
P
P
k&c----
20
,I’
0 CMCt
7
8
s
10
5SUBSTRATE
10CONCN.lmMl15
PH
Fig. I. Titration efficiency at various pH values for heptanoic and octanoic acid (5 and IO ,umoles).
Titrations were carried out in a mixture of IO mM of the corresponding lecithin, 0.1 M NaCl,
0.5 mM Tris, and IO mM CaCl,. This mixture is representative for an average assay. The amounts
of alkali required to reach a certain pH value in the presence of fatty acid were corrected for a
blank. The data are an average of three determinations.
Fig. 2. Michaelis curve of the hydrolysis of diheptanoyllecithin
with phospholipase A. Assay conditions: 0.5 mM Tris, 0.1 M NaCl, IO mM CaCI,, pH 6.5, and varying concentrations of substrate.
like sodium deoxycholate. Addition of this detergent had no influence on the rate of
breakdown. In the micellar region (above the CMC) Michaelis curves are obtained
which describe the progressive adsorption of the enzyme at the surface of the micelles.
Finally, a maximal rate of hydrolysis is obtained when virtually all enzyme molecules
are bound to the lipid-water interface.
Even at substrate concentrations below the CMC, where presumably mainly
monomers are present, some enzymatic activity could be demonstrated*. The enzyme activity, however, does not increase continuously with increasing monomer concentration. It reaches a maximum and then falls off with higher amounts of substrate,
finally giving a minimal activity at the CMC. In this monomeric region, however, a
poor proportionality was found between activity and enzyme concentration. Hence
evaluation of the kinetic constants umaxand K, is rather difficult. Work is in progress
to investigate by direct binding studies whether the enzyme is capable of binding
monomeric lecithin molecules. In any case, the observed activity against “monomers”
is low compared with the activity of phospholipase A against micelles. The preferential attack of substrate molecules present in micellar form is also shown in Fig. 3.
Enzyme activity is measured with two solutions of dihexanoyl-lecithin. Curve
A represents the slow hydrolysis rate of this substrate at a concentration just below
the CMC. Curve B with substrate concentration just above the CMC suggests that the
small amount of substrate present in micellar form is hydrolyzed at a much higher
rate. After about I min, however, most of the micelles are degraded and the remaining
substrate is hydrolyzed by the enzyme at the same low rate as found in curve A. From
these data it seems that phospholipase A has a pronounced preference to attack lecithin molecules present in certain organized structures. A similar behavior has been
reported earlier for phospholipase A from Crotalus adamante&.
also showed that the lipolytic enzyme,
ENTRESSANGLES
AND DESNUELLE~~
pancreatic lipase (EC 3.r.r.5)hydrolyzes short-chain triglycerides as emulsions and
“micellar” solutions at a much higher rate than the corresponding monomeric solu* Also dipentanoyllecithin
(with CMC greater than 60 mM) is very slowly hydrolyzed
enzyme in the concentration range where monomers are supposed to be present.
Biochim.
Biophys.
Acta,
by the
2x9 (1971) 252-266
G. H. DE HAAS
256
600
o
/
et al.
DIOCTANOYLLECITHIN
/
I-
AwONOMERS)
1oc,x)_o-o-~,o
DIHEPTANOYL‘LECITHIN
DIHEXANOYLLECITHIN
.9
_~~_~-~_J~%pOYL-
L
~IBSTRATE]
(m
M1
Fig. 3. Comparison of reaction velocities of the action of phospholipase A on monomeric lecithin
molecules and on lecithin micelles. Assay conditions: pH 7.0, 0.1 M NaCl, ro mM CaCl,. A, titration curve for hydrolysis of 13.5 mM dihexanoyllecithin monomers (just below CMC) ; B, titration
curve for hydrolysis of 14.5 mM dihexanoyllecithin (above the CMC, therefore a mixture of micelles and monomers). At the break point, after about I min, all micelles are hydrolyzed.
Fig. 4. Comparison of Michaelis curves of the hydrolysis of dinonanoyl-, dioctanoyl-, diheptanoyl-.
and dihexanoyllecithin
by phospholipase A. Assay conditions in all cases: pH 7.0, 0.1 M NaCl,
0.5 mM Tris, and I mM CaCl,. Substrate concentrations are expressed in mmoles/l substrate molecules present as micelles or liposomes respectively.
tion. Their finding that the enzyme displays a similar vmlLxfor the same substrate
present in “micelles” and in emulsion droplets might indicate a closely related architecture of both interfaces, which determines the rate limiting step in the hydrolysis.
In order to investigate the influence of enzyme activity on a gradually changing
lipid-water interface, a homologous series of short-chain lecithins was used as substrate. Light scattering experiments showed that the optically clear dihexanoyl- and
diheptanoyl-lecithin form spherical micelles containing about 25 and 40-50 monomers,
respectively*. The dioctanoyllecithin, however, gives a turbid solution in water, even
after ultrasonic treatment. This system is not a stable dispersion, but readily separates
into two layers, a very viscous lipid-rich lower phase and a clear supernatant containing the lecithin at a level of less than 0.3 mg/ml **. As judged from the high viscosity,
this lecithin produces probably large disc-shaped or elongated cylindrical micelles in
* Experimental details on the determination of micellar molecular weights of ionic detergents by
light scattering are described by HUISMANse. These measurements were carried out under similar
conditions of pH and ionic strength as used for the kinetic experiments.
* * X-ray investigation of the coacervate containing about 5% lipid showed the absence of lamellar and hexagonal structure. (Personal communication
from Miss A. TARDIEU, Laboratoire de
Genetique Physiologique, C.N.R.S., Gif-sur-Yvette,
France).
Biochim. Biophys. Acta, 239 (1971) 252-266
PHOSPHOLIPASE A
TABLE:
I
KINETIC
PARAMETERS
257
OF PANCREATIC
PHOSPHOLIPASE
A
WITH
SHORT-CHAIN
LECITHINS
AS
SUB-
STRATES
Assays were performed as described under methods. The assay conditions (pII 7.0, I mM CaCI,,
and 0.1: M NaCl) for the three substrates were arbitrarily
chosen, but are representative for the
comparison of the kinetic parameters. rmax and X, * values were obtained from Lineweaver-Burk
plots. urnaxis expressed inpmoles/mg per min. K, is expressed as concentration of substrate present
in the form of micelles.
Substrate
vmax (at [Ca2+] = I mM)
Dihexanoyllecithin
IO
Diheptanoyllecithin
szl
Dioctano~lleci~in
_...
* K, represents the dissociation
later.
KS (mM)
6.3
4.7
7.4
--.
constant of the enzyme-substrate
complex
as will be explained
water which are insoluble and in equilibrium with monomeric species. Dinonanoyllecithin is the first of the higher homologues of lecithin which form liposomes in water
as could be demonstrated by phase-contrast microscopy*.
As shown in Fig. 4, large differences in enzyme activity exist even for these
chemically very related substrates. As was anticipated, liposomes of dinonanoyllecithin are not degraded by the enzyme. In order to compare the enzyme action
against the micelles of the lower homologues, the molar substrate concentrations
plotted on the abscissa have been corrected for the amount of substrate present as
monomers (CMC). The assumption is made that in these micelles all molecules are
exposed to the solvent and in principle are capable of reacting with the enzyme. In
this way the main difference with a true solution is the regular organization of the
moiecules in a lipid-water interface. No corrections were made for the low enzyme
activity toward monomers.
By converting the Michaelis curves of Fig. 4 into Lineweaver-Burk plot+,
the kinetic parameters of the enzyme with the various substrates are obtained (see
TABLE I). The K, values * * which are expressed in M substrate entirely present in the
micellar state show roughly a constant value for the three substrates. However, since
micellar size and hence the number of molecules per micelle and area per molecule
change, features which have been neglected in our assumption, a direct comparison
of the K, values is not justified. TABLE I shows an 8o-fold increase in ZI,,, (at
[CaZ+] = 10-a M) going from dihexanoyl- to dioctanoyllecithin. Apparently the differences in architecture of the micellar structures of these related lecithins have a very
large bearing on subsequent reaction steps during catalysis. If we assume a classical
acylation-deacylation
mechanism also for this special hydrolase, we are forced to
conclude that at least the rate-limiting step must be strongly influenced by one (or
more) of the parameters of the interface. An attractive hypothesis which was originally
suggested by HUGHEF and by SHAH AND SCHULMAN~~ might be to consider the area
per molecule of substrate in the interface (or the charge density per unit of surface) as
one of the most important factors which govern the rate of hydrolysis. Calculations of
the area per molecule for spherical micelles of dihexanoyl- and diheptanoyllecithin
* Observations were made under a light microscope fitted with crossed nicol prisms and a firstorder red compensato+8.
* + In the next section it is made plausible that this dissociation constant is indicative of the true
equilibrium constant of the enzyme-substrate
complex.
&whim.
Biophys.
A&,
239
(1971)
252-266
258
G. H. DE
HAAS
et al.
approximate values of 125 i%a and roe AZ, respectively. Although the exact geometry of dioctanoyllecithin micelles is still unknown, it seems to be reasonable to
attribute an area of approximately 90 Aa per molecule to this, still micellar, interface,
taking into account the expected close packing of dinonanoyllecithin molecules in
liposomes with an area per molecule of approximately 75-So As.
The slow hydrolysis by the enzyme of dihexanoyl- and diheptanoyllecitllin
micelles and the high susceptibility of the dioctanoyllecithin micelles together with
the observed inertness of the dinonanoyllecithin liposomal structure might indicate
that an area per lecithin molecule in the interface of about 90 A2 provides a highly
favorable packing of the molecules for the enzyme. In this respect it should be remarked that the enzymatic degradation of dioctanoyllecithin spread as a monomolecular
layer45 shows a pronounced maximum at a pressure of 8 dynes/cm, which is equivalent
to an interfacial area per molecule of approximately 90 A*. Also monolayer studies of
COLACICCO
AND RAPPORTS’ who investigated the breakdown of natural, long-chain
lecithins by two different snake venom phosphohpases A, demonstrated an optimal
action of the enzymes at a surface pressure of 12 dynes/cm, which also corresponds
with an area per molecule of about go A2.
give
Kinetic parameters of pancreatic phos$holipase A with short-chain lecithins as substrate.
Proposed mechanism
The pancreatic enzyme is completely inactive in the standard assay system if
Ca2+ is omitted’*. Fig. 5 shows that the enzyme absolutely requires Ca2+ at all
pH values studied, suggesting a role of this metal ion in the fixation of phospholipase
A to its substrates and/or in the reaction steps of the catalytic process itself. As will
be shown in a subsequent paper, the requirement of the enzyme for Ca2+ is absolutely
specific and no other common metal ion can substitute for it*. BaZf and SP+ are pure
__o.-_---o-
PHBO
_-o-
PHQO
with
Fig. 5. Effect of pH on the requirement for Ca s+ in the hydrolysis of diheptanoyllecithin
phospholip~e
A. Assay conditions : 0.5 m&l Tris, 0.1 M NaCf, r4 mM diheptanoylIec~thin, and
varying CaCl, concentrations. Measurements were performed at pH values as indicated.
* The purified enzyme contained a low amount of calcium as measured by titration with murexideaa. This calcium was mainly introduced during the dialysis against distilled water. No indication for the presence of other metal ions was obtained.
Biochim.
Biophys.
Acta,
2x9 (1971)
252-266
259
PHOSPHOLIPASE A
competitive inhibitors for phospholipase A, in other words they are able to bind to
the enzyme with a dissociation constant close to that of Ca2+. However, the enzymemetal-substrate complex formed did not lead to product formation.
Mg2+ is unable to bind to the enzyme and its presence has no influence on enzyme activity measured with Ca2+.
What is the pathway leading to the formation of the active enzyme-metalsubstrate complex ? In order to investigate more closely the possible pathways, initial
reaction rates of the enzyme with diheptanoyllecithin were measured as a function of
varying substrate and Ca2+ concentrations and pH.
Fig. 6a shows at pH 7.0 the double reciprocal plots of enzyme activity versus
substrate concentrations for various fixed concentrations of CaCl,. The expermental
points fit straight lines which gave a common intersection on the abscissa after
extrapolation. This intersection was used to calculate the dissociation constant K, of
the enzyme-substrate complex. In Fig. 6b, at the same pH, double reciprocal plots of
enzyme activity versus Ca2+ concentrations are given for various fixed substrate concentrations. Again straight lines were obtained with a common intersection on the
X-axis which was used to calculate the dissociation-constant Kca2+ of the enzymemetal complex. Similar Lineweaver-Burk patterns with common intersections on the
abscissa were obtained at various pH values between 5.0 and 9.0, indicating an identical mechanism for the formation of the active enzyme-metal-substrate
complex in
the pH range studied. ROHOLT AND SCHLAMOWITZ 26discussedtwo ordered pathways
being both in agreement with their inhibition experiments with Ba2+. In the first
pathway, the enzyme binds first reversibly with Ca2+, and the enzyme-Ca2+ complex
forms subsequently the active enzyme-metal-substrate complex:
S
E+Ca2+ + EsCa2+ + E .Ca2+.S
(I)
According to the second mechanism, there is first a formation of an enzyme-substrate
complex which then reacts further with Ca2+ to form the enzyme-metal-substrate
complex :
Cae+
E-t-S
+ E.S
+
E.Ca2+.S
(4
A third possibility is that not the free lecithin, but a Ca2+-lecithin complex is the true
substrate for the enzyme; but this was rejected by these authors as being not in agreement with their kinetic data. Our failure to demonstrate any binding of Ca2+with the
short-chain synthetic lecithins forced us to consider the last mechanism also as
extremely improbable.
Direct binding studies in our laboratory, the result of which will be reported in
a subsequent paper, are not in disagreement with the ordered mechanisms (I) and
(2).In other words, the enzyme is indeed capable of binding Ca2+ in the absence of
substrate, and the enzyme can form a complex with lecithins without the presence of
Ca2+. However, the kinetic results as shown in Figs. 6a and 6b do not support an
ordered mechanism. According to pathway (I), one would expect a reaction velocity
for infinite substrate concentration which is independent of the Ca2+ concentration.
In other terms, Lineweaver-Burk plots as shown in Fig. 6a should give a common
intersection on the ordinate, which is clearly not true in this case. The second pathway (2) is characterized by a reaction velocity which becomes independent of substraB&him.
Biophys.
Acta,
239
(1971)
252-266
G. H. DE HAAS & ai.
“‘d
b
Fig. 6 (a) Lineweaver-Burk
plots of hydrolysis of diheptanoyllecithin
by phospholipase A at pH
7.0 with varying Ca*+ concentrations.
Assay conditions: 0.5 m&I Tris or acetate buffer, 0.1 5%
NaCl, varying substrate concentrations, and 5 different CaCI, concentrations as indicated. The
dotted line for [Ca*+] = infinite was obtained from Fig. 6b by plotting the I /Y’~~~ values versus
I/[.‘?]. The intersection of the dotted line with the ordinate represents r~vmax at infinite Ca’+
concentration. [S] is expressed as moles/lof lecithin present in the form of micelles. Similar patterns
were obtained for all pH values between 5 and 9. (b) Double reciprocal plots of reaction velocities
and Ca*+ concentrations obtained from hydrolysis of varying substrate concentrations (diheptanoyllecithin) by phospholipase A. Assay conditions: the same as in (a). The dotted line for [S] = infinite
is obtained from (a) by plotting the I /vmax values versus the reciprocal of the CaBf concentration.
The intersection of the dotted line with the ordinate represents I /r~‘~s~ at infinite substrate concentration and equals I /vmax at infinite Ca2f concentration (see a). Similar patterns were obtained
for all pH values between pH 5 and pH 9.
te concentration
at infinite Ca*+ concentration.
Plots of this type (see Fig. 6b) should
give again a common intersection on the ordinate which is apparently not the case.
Figs. 6a and 6b do show a common intersection on the abscissa which can be expected
for a random mechanism in which the metal ion acts by combining with the enzyme
Bioshim.
Biophys.
Acta,
239 (1971) 252-266
PHOSPHOLXPASE
261
A
independently
of the substrate
dently of the Ca e+ concentration
and the substrate
:
combines
with the enzyme
indepen-
E@@@+ca2+\\
F.c&+.S
(3)
_
\,.,/
on
In this case, described by DIXON AND WEBBED as Case I, the common intersection
the abscissa (Fig. 6a) allows us to calculate the K, value, the dissociation constant of
the E-S complex, whereas Kca3+, the dissociation
constant
of the enzyme-metal
complex, directly follows from the intersection
in Fig. 6b.
The pH dependence of the kinetic parameters K, and Kcaa* obtained from Figs.
6a and 6b at various pH values is shown in Fig. 7. Although the pH-stat assay technique did not allow us to measure enzyme affinity at pH values below 5.0, it seems that
optimal binding between enzyme and substrate takes place at slightly acidic conditions. Direct binding experiments
between enzyme and shortchain
lecithins, which
will be reported later, indicate a binding constant at pH 4.0 which is very close to the
104xM
e
140
120
.
~
"MAX!
I
100
400
1
I
80
200 ’
1
4
5
6
7
8
9
10 pti
3
4
5
6
7
8
QPH
Fig. 7. Dependency of KS and Kc gs + on pH for hydrolysis of diheptanoyllecithin with phospholipase A. K, and Kcszc values were obtained from Figs. 6a and 6b measured at pH values between
5 and 9. At all pH values common intersections with the abscissa were observed in both double
reciprocal plots. K, is expressed in moles/l of substrate in micellar form (K, varies between 1.3
and 16 mM (o), and Kc8ai between 2.6 and 0.25 mM {o).
Fig. 8. pH optimum of pancreatic phospholipase A hydrolysis of diheptanoyliecithin. Maximal
velocities were obtained from Figs. 6a and 6b at infinite substrate and Car+ concentration. Assay
conditions as described in Fig. 6a. The values below pH 5 were obtained by the hydroxamate
method at a fixed high Cae+ concentration.
Biochim.
Biophys.
Acta,
239 (1971) 252-266
G. H. DE HAAS
262
et al.
values found at pH 5.0 and 5.5. Above pH 5.5 one observes a rapid decrease in binding
efficiency. As regards the binding of Ca”+ by the enzyme, it is obvious that slightly
alkaline conditions favor this association. At pH values below 6.0, a rapid decrease in
affinity between metal and enzyme can be observed. Again a very similar pH dependence of the phospholipase A-Ca2+ binding was found by direct binding studies.
From the experiments described in Figs. 6a and 6b, which were carried out
at nine different pH values between 5.0 and 9.0, maximal velocities for infinite Caz+
and infinite substrate concentrations are obtained in the following way: the intersections on the ordinate in Fig. 6a represent the reciprocal values of maximal velocity
for infinite substrate, but defined Ca2+ concentration. These values were plotted ve~s’s21s
the reciprocal Ca2+ concentration (dotted line in Fig 6b). The intersection with the
ordinate then gives the reciprocal value for ZImaixat infinite substrate and Ca2+ concentration. The corresponding intersections with the ordinate in Fig. 6b can be transformed into the dotted line in Fig. 6a. The intersection of this line with the ordinate
([S] and [CaZ+] = infinite). Fig. 8 shows the
also gives the reciprocal value of tiUmax
pH dependence of the so-obtained vmaxvalues. Below pH 5.0 enzymatic activities had
to be measured by the hydroxamate method. Although various substrate concentrations were used, the assays were performed at only one high Ca2+ concentration. The
obtained extrapolated vmaxv alues are therefore dotted, but are indicative of a rapid
decrease of the amax at infinite Ca 2+ concentration at these pH values. With these
short-chain lecithins as substrate, phospholipase A has a pronounced pH optimum
between pH 5.5 and 6.0. The same pH optimum of about pH 6 was also obtained
by monolayer techniques with dioctanoyllecithin as substrate.
From this result one might speculate that for the decomposition of the metalenzyme-substrate complex at least one amino acid side chain is important which
should be in unprotonated form, with a pK value around 4.75 (presumably a carboxyl
group).
The pH optimum of pancreatic phospholipase A in vitro is generally found to
be around pH 8.0 It cannot be precluded, however, that under the conditions of
these measurements an apparent pH optimum is found due to the fact that bile salts
loose their emulsifying action at more acidic pH values.
InflzLence of high NaCl concentrations on the activity of phospholifase
A with short-chain
lecithins as szlbstrate.
As was shown in Fig. 4, the enzyme activity against micellar interfaces of closely related lecithins is strongly dependent on the chain-length of the substrate. We
supposed that small differences in the architecture of the lipid-water interfaces, for
instance the area per molecule, are responsible for these large effects on enzyme activity. That the relatively low activity of the enzyme zlersztsdihexanoyl- and dihept~oyllecithin compared with the dioctanoyl derivative is probably caused by physicochemical differences of the interfaces, is supported by the finding that high ionic
strength enhanced reaction velocities.
As shown in Fig. 9, the addition of large amounts of NaCl to aqueous solutions
of diheptanoyllecithin not only lowers considerably the CMC, but also increases
dramatically the activity of the enzyme at the same time. A similar activating effect
was noted with dihexanoyl- and dioctanoyllecithin as substrate. Maximal velocities
at infinite substrate concentration ([Cazf] = I mM) were obtained from LineweaverBiochinz.Biophys. Acta, 239
(1971) 252-266
PHOSPHOLIPASE
500
263
A
39 M NaCl
1
400
/
6
I
UlAX
%lAX
1000
5OOOi
800
600
400
-o-o__4-o-o-0
0
24
6
[SUBSTRATE]
8
200
1M NaCl
IO
0
12
M
NaCL
a
14
( mM)
Fig. 9. Influence of varying NaCl concentrations
pholipase A. Assay conditions: 0.5 mM Tris, I
All measurements were performed at pH 7.0.
Substrate concentrations are given in moles/l of
servation of the change in CMC with increasing
on the hydroiysis of diheptanoyllecithin by phosmM CaCI,, and varying substrate concentrations.
NaCl concentrations are indicated in the figure,
lecithin (micellesplus monomers), which allows obsalt concentration.
Fig. IO. Plot of z+naxvalues zleysplsNaCl concentration obtained from hydrolysis measurements of
diheptanoyl- and dioctanoyllecithin
by phospholipase A. Assay conditions as described in Fig. g.
vmsx (at [Ca2+] = I mM) values were obtained from Lineweaver-Burk plots in which [S] was
corrected for the CMC. Abscissa at the left gives the scale for dioctanoyllecithin
and abscissa at
the right that for diheptanoyllecithin.
Insert: log rmax values plotted ZJWSUS
NaCl concentration.
mitx with increasing salt concentration.
Straight lines indicate an exponential increase of the z1
DHL = diheptanoyllecithin;
DOL = dioctanoyllecithin.
Burk plots of the curves from Fig. g. A plot of these umsx vahres as a function of the
NaCl concentration
is given in Fig. IO for two substrates.
For both compounds an
exponential increase in activity was observed and no limit could be reached. Compare
the insert in Fig. ro where a linear relationship
exists between log zfmsXand the NaCl
concentration.
One may wonder whether this salt effect is primarily caused by alteration in enzyme conformation
and activity or, which is more plausible, by changes in
the lipid-water
interfaces. Kinetic measurements
of phospholipase
A on monolayers
of dioctanoyllecithin
at a fixed low pressure showed no such increase in activity
whether NaCl (2 M) was present or not. Since high salt concentration
does not influence the lecithin monolayer, these results seem toindicate that the high ionic strength
does not strongly modify the enzyme structure or properties.
Finally Table II shows the salt dependence on the kinetic parameters K, and
Kcaa+. Small differences in enzyme conformation
caused by dehydration
might be
responsible for the slightly higher affinity between enzyme and substrate
(K,) at
increasing salt concentrations.
Taking into account the fact that generally purely
ionic interactions
will decrease in solutions of increasing ionic strength (cf. the raise of
Kca2+ with increasing NaCl concentrations),
the small shift in K, values might indicate
_%~hi?% Biophys. Acta,
239
(1971)
252-266
264
G. H. DE HAAS
TABLE
INFLUENCE
WITH
et al.
II
OFINCREASING
A SHORT-CHAIN
NaCl
LECITHIN
CONCENTRATIONS
AS
ON
KINETIC
PARAMETERS
OF
PHOSPHOLIPASE
A
SUBSTRATE
Assays were performed at pH 7.0 (0.5 mM Tris). Values for ZJ
mex (pmoles/mg per min) and K,
(mM) were obtained from Lineweaver-Burk
plots with diheptanoyllecithin as substrate at varying
CaCl, concentrations. K, is expressed in substrate concentration present in the form of micelles.
Kcaa+ values were obtained from double reciprocal plots of activity uwsus CaP+ concentration at 5
different substrate concentrations. At every NaCl concentration a common intersection on the
abscissa was obtained.
______~~
NaCl Concn. (M)
urnax(at [Ca2+] = co)
KS (mM)
KcP
(mM)
53.5
322
0
I
2
1600
3
4300
4.6
4.4
2.3
IO
1.0
IO
0.2
2
_
a considerable contribution of the Van der Waals interactions forces to the binding
energy between enzyme and substrate *. It is difficult to provide an explanation for
the 80 fold increase in vmax ([Caz+] and [S] = infinite) in the range from o to 3 M
NaCl. These high salt concentrations will undoubtedly change the parameters of both
the lipid-water interface and the solvent. For instance, the water structure in and
surrounding the micellar surface will be altered, the charge-charge interactions in the
interface due to the polar head groups will diminish, and the dielectric constant of the
solvent will be considerably lowered. More physicochemical studies on the salt-induced
changes of such interfaces have to be done before a correlation with enzyme activity
can be made.
On the other hand, from a practical point of view, the salt induced increase in
activity might be useful to extend our knowledge about other lipolytic enzymes. Not
only the pancreatic phospholipase A becomes extremely active at high ionic strength,
but also, for instance, the action of pancreatic lipase on lecithins is greatly enhanced.
This enzyme which specifically cleaves fatty acid ester bonds in the r-position of
phospholipids in a non-stereospecific way has been used for preparative purposes to
obtain 2-acyllysolecithins ho.However, its action on phospholipids without the addition
of salt is slow, and the prolonged incubation conditions needed for that reason may
cause acyl migration. In the presence of high salt concentrations, lecithins are degraded very effectively by lipase and the enzymatic hydrolysis can be followed by pH-stat
techniques.
CONCLUSIONS
Although pancreatic phospholipase A is able to hydrolyze slowly substrate
molecules in monomeric dispersions, the enzyme attacks substrate molecules which
are organized in certain lipid-water interfaces at a much higher rate. Experimental
difficulties encountered in the kinetic measurements in the monomer region prevent
us so far from explaining the differences in terms of binding (KS) or catalysis (urnax).
Slightly different substrates, such as a homologous series of lecithins, produce different lipid-water interfaces (size of the micelle, area per molecule in the interface, etc.).
* It should be remarked once more, however, that comparison
ent interfaces is not justified.
Biochim.
Biophys.
Ada,
239 (1971)
252-266
of K, values obtained with differ-
PHOSPHOLIPASE
A
265
Therefore the affinity constants between enzyme and even very similar substrates
cannot be compared with each other. The observed large differences in the ZI,,,
values with such substrates cannot be explained by differences in chemical reactivity.
One of the interfacial parameters which might govern the maximal rate of hydrolysis
is the area per molecule, or charge density in the lipid-water interface.
The enzyme absolutely requires Ca 2+ for activity and this generally found
characteristic property has been explained in terms of a specific bridge of the metal
between the enzyme and an anionic site of the substrate&l. Although such a bridge
mechanism cannot be precluded for more acidic phospholipids, the kinetic results
obtained with lecithins under the conditions used do not support this function of the
metal and rather point to a mutually independent binding of metal and substrate to
the enzyme. In this respect it should be remarked that direct binding experiments
failed to demonstrate any affinity of Ca 2+ for short-chain lecithins, which is in agreement with the results of ROJAS ANDTOBIAS~~and of HAWSERANDDAWSON~~.
The fundamental role of certain interfacial parameters in the highly effective
hydrolysis of lipids by lipolytic enzymes is still far from being understood. It is hoped
that a simultaneous investigation of the enzyme kinetics by “bulk” methods and
monolayer techniques will aid in a better understanding of these Iipoprotein interactions.
ADDENDUM
A#mximated
calcdations of the nrea/molecule in miceiles of short-chain Lecithins
Using the spherical micelle model (Harkin’s micelle) and the micellar molecular
weight as determined by light scattering, the surface area occupied by each lecithin
molecule can be calculated as follows: Light-scattering data indicated that micelles of
dihexanoyllecithin at 0.1 M NaCl at 25’ consist of 25 monomers. With the volumes of
the polar headgroup (336 A3), the CH,-group (54 A3) and the CH,-group (27 A3) as
determined by GULIK-KRZYWICKI et al .44, the total volume of the spherical micelle,
consisting of 25 monomers (16500 A3) enables us to calculate the radius of the micelle
to be 15.18 A. From the total surface of the micelle (4~ r2 = 3140 ifa) and the number
of monomers (25) per micelle, the area occupied by one dihexanoy~ecithin molecule
is found to be 125 AZ. A similar calculation for dihept~oyllecithin
(about 45 monomers per micelle) gives an area per moIecule of 100 AS.
ACKNOWLEDGEMENTS
Thanks are due to Mrs. G. J. Burbach-Westerhuis and Mrs. M. C. M. H. de
Waal-Bouman for their skilful technical assistance. The authors are much indebted to
Professor G. Zografi (Ann Arbor, U.S.A.) for the infomration on monolayer experiments prior to publication and for assistance in the preparation of the manuscript.
The stimulating critica discussions with Professor M. Lazdunski and with Dr. R.
Verger (Marseille, France) are highly appreciated. The authors thank Drs. R. J. M.
Tausk (Van ‘t Hoff Laboratory, State University, Utrecht) for giving information on
the ultracentrifugation runs and light scattering experiments prior to publication.
266
G. H. DE HAAS et al.
REFERENCES
I L. L. M. VAN DEENEN AND G. H. DE HAAS,Biochim. Biophys. Ada, 70 (1963) 538.
z J.C. VIRAL AND A. 0. M. STOPPANI, Rev. Sot. Arg. Biol., 43 (1967) 199 (Chem. Abstr., 71 (1969)
9821).
3 B. T. CURRIE,D. E. OAKLEY AKD C. A. BROOMFIELD,h'ature,
220 (1968)371,
4 J,LSALACH, P. TURIKI,J.HAUBER,R.SENG,H.TISDALEAND T.P.%~~~~,I?iochern.
Biophys.
Res. Commun., 33 (1968)
936.
5 M. A. WELLS AND D. J.HANAHAN, Biochemistry, 8 (1969) 4’4,
6 T. W. Wu AND D. 0. TINKER,Biochemistry, 8 (1969) 1558.
7 B.M. BRAGANCA,Y.M.SANRRAYAND
R.C. GHAIXALLY,Toxicon,7(1g69)
151.
8 J. M. AUGUSTYN AND W. B. ELLIOTT, Biochim. Biophys. Acta, 206 (1970) 98.
9 D. N. SAKHBIOV,V.M. SOROKIN AND L.YA.YUKELSON, Biokhimiya, 35 (1970) 13,
IO A. T. Tu, R. B. PASSEY AND P. M. TOOM, Arch. B&hem. Biophys., 140 (1970) 96.
II A. WAHLSTR~~M,To&con, g (1971) 45.
12 E. HABERDIANN AND K. G. REIZ,B~oekem Z., 34.3 (1965)192.
13 S. A. BARKER, A. W. MITCHELL,K. W. WALTON AND P. D. WESTON, Clin. Chim. Acta, 13
(1966) 58~.
14 G. H. DE HAAS,N. M. POSTEMA,
~.NIEUWENHUIZENAND L.L.M. VAN DEENEN, Biochim.
Biophys. Ada, 159 (1968) 103.
15 G. H. DE HAAS,N.M.POSTEMA,W.MIEUWENHUZEN
AND L.L.M.VAN DEENEN, Biochim.
Biophys. Acta, 159 (1968)118.
16 W. L. MAGEE, J. GALLAI-HATCHARD,H. SANDERS AND R. H.S. THOMPSON, Biochem.J., 82
(1962) ‘7.
17 K. A. WITTICH AND H. SCHMIDT,Enzym.BioZ. Cl&.,IO (1969)
477.
18 C. FIGARELLA,F.CLEMENTE AND 0. GUY, B~ocke~~.B~ophys.
Acta, 227 (1971)213.
19 B. ARNESJZI,J.BARROWMAWAND
B.BoRGsTR~~~,~~~~C~~~.S~~~~.,~I
(1967)2897.
20 G. Ii.DE HAAS, A. J.SLOTBOOM, P. P.M. BONSEN, L. L.M.VAN DEENEN, S.MAROUX, A.
PUIGSERVER AND P. DESNUELLE,Biochim.Biophys.
Acta, 221 (1970) 31.
21 G. H. DE HAAS, A. J. SLOTBOOM, P. P. M. BONSEN, W. NIEUWENHUIZEN, L. L. M. VAN
DEENEN,S.MAROUX,V. DLOUHAAND P.DESNUELLE, Biochim.Biophys.Acta.
221 (1970)54.
22 Y. SAMEJIMA,S. IWANAGA, T. SUZUKI AND S. KAWAUCHI, Biochim.Biophys.
Acta, 221 (1970)
417.
23 W. SHANKLAND, Chem. Phys. Lipids, 4 (1970) rog.
24 R. M. C. DAWSON, B&hem.
f., 98 (1965)
35~.
25 G. H. DE HAAS, C. H. T. HEEMSKERK, L. L. M VAN DEENEN, R. W. R. BAKER, J.GALL.AIHATCHARD,~. L.MAGEE AND R. H. S.THOMPSON,
in A.C. F~~~~~,Biochemical
Problems of
Lipids, Vol.I,Elsevier,
Amsterdam, x963,p. 224.
26 0. A. ROHOLT AND M. SCHLAMOWITZ,Arch.Biochem.Biophys.,
g4(1g61) 364.
27 E. CUBERO ROBLESAND H. DE IONGH,Rec. Trav.Chim., 86(1g67) 762.
28 E. CIJBEROROBLES AND D.VAN DEN B~~~,Biochim.
Biophys.Acta,
187(1969)52o.
29 J. AUG~JSTYNAND W. B. ELLIOTT,AnaZ.Biochem., 31 (1969)246.
30 P. BECHER, J. Phys. Chem., 66 (1962) 374.
31 B. ENTRESSANGLES AND P.DESNUELLE, Biockim, Riophys.
Acta, 159(1968)28j.
32 H. F. HUES&IAN,
Proc. Kon. Ned. Akad. Wetenschap. Ser. B., 67 (1964) 367.
33 it. D. BANGHAM, in J. A. V. BUTLER AND D. NOBLE, Progress in Biophysics and ~~olee~~la~
Biology, Pergamon,Oxfordand New York, 1968,p. 29.
34 H. LINEWEAVER AN~DD. BuKK,J. Am.Chem. Sot., 56(x939)658.
35 A. HUGHES,
Biochem. J., 29 (1935) 437.
36 D. 0. SHAH APZDJ,H. SCHULMAN, J. Colloid Interface Sci., 25 (1967) 107.
37 G. COLACICCO AND M.M. RAPPORT,J. LipidRes.,
7 (1966)258.
38 T. ONISHIAWD S. EBASHI,J. Biochem. Tokyo, 54 (1963) 506.
39 M.DIXON AND E.C. W~~~,Enzymes,AcademicPress,NewYork,znd
cd.,1964,P.429.
4o A. J. SLOTBOOM,
G. H. DE HAAS,
G. J.BURBACFI-WESTERHUISAND L. L.M.VAN DEENEN,
Chem. Phys. Lipids, 4 (1970)30.
41 R. M. C. DAWSON, Natzkve,
196 (1962)67.
42 E. ROJAS AND J.M. TOBIAS,Biochim. Biophys. Acta, g4 (1965) 395.
43 H. HAWSER AND R. M. C. DAWSON, Ewopean J. Biochem., I (1967) 61.
44 T. GULIK-KRZYWICKI,E. RIVAS AND V. LUZZATI,J. Mol.Biol.,
27 (1967) 302.
45 G. ZOGRAFI,to be published.
Biochim. Biophys. Acta, 239 (1971)
252-266